Skip to content

Instantly share code, notes, and snippets.

@philwinkle
Created August 20, 2013 01:21
Show Gist options
  • Save philwinkle/6276076 to your computer and use it in GitHub Desktop.
Save philwinkle/6276076 to your computer and use it in GitHub Desktop.
# So, you forgot to use logrotate and now you're staring at a 30GB system
# log on your production server.
# Never fear...
# Tail an arbitrarily large number of bytes off the end of the thing.
# This example takes approximately the last 5GB.
tail -c 5000000000 system.log > ~/system.log.tmp
# But, we only want the last month of log data. Let's find the line number
# where this month's reporting starts.
grep -m 1 -n 2013-08- system.log.tmp
# Pretend that the line number you got back was 11108105.
# You also need to know the total length of the file.
wc -l system.log.tmp
# Again, pretend the total file length was 12208454 lines. Now take the difference.
12208454 - 11108105 = 1100349
# So, to get just the last month's data, we tail one last time.
# Note that I've incremented the difference I calculated by 1.
tail -n 1100350 system.log.tmp > system.log.2.tmp
# Now if only we could find every *unique* error, exception, warning, and notice
# in the last month's data. Oh wait, we can...
egrep 'Warning:|Notice:|Error:|Exception:' system.log.2.tmp | cut -c 36- | sort | uniq > system.log.3.tmp
# Cheers :D - NWJ
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment